advert

JOURNAL OF RESEARCH IN NATIONAL DEVELOPMENT VOLUME 8 NO 2, DECEMBER, 2010


 

 

INTERFERENCE MITIGATION TECHNIQUES IN WIRELESS COMMUNICATIONS SYSTEMS

 

J.A. Oyedepo, Y.O. Salihu and Adenike Folaponmile 

Computer Engineering Department, Kaduna Polytechnic, Kaduna, Nigeria

E-mail:nikkyfola04@yahoo.com

 

Abstract

Co-Channel Interference, Intersymbol interference and fading are major impairment to the high-capacity transmission in power- and band-limited wireless communication channel. This paper presents an overview of interference mitigation techniques in wireless communications systems. Linear filtering, equalization, and diversity combining techniques are some of the interference mitigation techniques described in this paper.

 

Keywords: Wireless communications, filtering, equalization, diversity, multiuser detection

 


Introduction

Within the last two decades, wireless communications has evolved from an optional convenience to an indispensable necessity of daily life. Advances in digital signal processing, digital computing and radio transmission technologies have facilitated the introduction of a wide range of wireless communication service ranging from second generation mobile communications system such as GSM, IS-95 and PDC which provide narrowband communications links mostly for voice and text traffics with high mobility to high speed private and public access wireless local/personal area networks (WLAN/WPAN) such as Wi-Fi and Bluetooth which deliver broadband multimedia services with limited mobility.

 

The term wireless communications refers to transfer of information via electromagnetic or acoustic waves over atmospheric space rather than along a cable (Lee, 1982). The apparent wrinkle between such a scheme and conventional wired systems is the presence of the wireless channel as the medium over which the communication must take place. Unfortunately, more often than not, this medium is hostile in regards to attenuating, delaying and even completely distorting the transmitted signal. The wireless medium introduces difficulties for communication by its inherent nature.

 

In wireless mobile communications, the transmitted signal is subject to various impairments caused by the transmission medium combined with the mobility of transmitters and/or receivers. Because of the randomness of the mobile propagation channels and limited radio spectrum, co-channel interference (CCI), fading and Intersymbol Interference (ISI) are the major impediments to high-capacity transmission in power and bandwidth-limited wireless communications systems.

Fading is counter measured by channel coding and interleaving techniques such as transmit / receive antenna diversity schemes. ISI from multi-path reception can be combated by various linear / nonlinear type equalization techniques employing symbol-by-symbol detection methods such as decision feedback equalizer (DFE) or sequence-estimation methods such as maximum likelihood sequence estimation (MLSE). In cellular networks, CCI is the interference from neighboring cells using the same radio channels. As the frequency reuse factor decreases to one, CCI becomes unavoidable due to the channel reuse in adjacent cells. CCI is best handled by the joint MLSE receiver. Less complex linear filter type receivers suppresses CCI by controlling the filter coefficients in the sense of maximizing the signal-to-interference-plus-noise-ratio (Liang et al, 1996).

In typical wireless mobile communication systems, CCI, ISI and fading often arise together. Hence, the receiver designs for mitigating these impairments in joint fashions are quite common. In filter-based approaches, CCI is suppressed by a feedforward linear filter while ISI is mitigated by a concatenated decision feedback filter (Liang et al, 1996). For the receivers with multiple antennas, diversity combining techniques broaden the freedom in interference mitigation receiver designs. MLSE or DFE type receivers, combined with diversity combining combat ISI and fading jointly. For the suppression of CCI in flat fading channels, an optimum linear minimum mean square error (MMSE) combining technique was suggested by (Winters, 1984). Multiuser detection (MUD) algorithms detect all co-channel signals unlike the filter- based methods treating all co-channel signals, except the desired one, as interference

 

Interfernce in wireless communications

Propagation channels

The wireless communications channel suffers from many impairments such as the thermal noise often modeled as Additive White Gaussian Noise (AWGN), the path loss in the power as the radio signal propagates, the shadowing due to the presence of fixed obstacles in the radio path, and the fading which combines the effects of multiple propagation paths, and the rapid movement of mobile units reflectors. Upon the signal transmission, different signal copies undergo different attenuation, distortion, delays and phase shift. The overall performance of the system can be severely degraded a result of these problems.

 

In a fixed environment, fading is affected by changes in atmospheric conditions such as rainfall. But in a mobile environment, where one of the two antennas is moving relative to the other, the relative location of various obstacles changes over time creating complex transmission effects.  Slow fading occurs when the coherence time of the channel is large relative to the delay constraint of the channel. The amplitude and phase change imposed by the channel can be considered roughly constant over the period of use. Slow fading can be caused by events such as shadowing, where a large obstruction such as a hill or large building obscures the main signal path between the transmitter and the receiver. Power control techniques are often used to combat the effects of shadowing (Stuber, 2001).

 

 Fast fading often referred to as multipath fading, occurs when the coherence time of the channel is small relative to the delay constraint of the channel. The amplitude and phase change imposed by the channel varies considerably over the period of use. Diversity and coding techniques are often used to combat multipath fading (Stuber, 2001).

 

Intersymbol interference

In radio channels for digital communications, Intersymbol Interference (ISI) is one of the main causes of system performance degradation. ISI is due to multipath propagation when the delay spread of the channel is large compared to the duration of modulated symbol (Qureshi, 1985). ISI can also be introduced by an over transmit and receive filter response that is not a Nyquist pulse. The ISI result in non-flat transfer functions in frequency domain such that all the frequency components in the transmitted signal may not experience similar amplitude and phase variations. An equalizer is a digital filter used in digital communications to correct or equalize ISI. Two types of equalizers are used to mitigate ISI by using linear or nonlinear techniques: symbol-by- symbol equalization such as Decision Feedback Estimation (DFE) and sequence estimation such as MLSE (3GPP, 2005).

 

Co-channel and adjacent-channel interference (CCI and ACI)

CCI is introduced when a frequency band is shared by multiple users at the same time. In cellular systems, CCI arises by the frequency reuse in neighboring cells. As frequency reuse factor decreases in order to increase the system capacity, CCI increases. It also increases as the distance between the co-channel cells decreases. Therefore, the performance of a frequency reuse system is limited by CCI rather than by additive noise (IEEE, 2003). For wireless communication systems such as EDGE, which uses smaller cell size and an aggressive frequency reuse strategy, CCI mitigation is an important issue for spectral efficiency increase. Diversity techniques and multiuser detection (MUD) approaches are effective ways for mitigating CCI (Laster and Reed, 1997).

 

Similarly, in ACI, the signals received from one channel but smeared into adjacent channels due to imperfect receive filtering or imperfect frequency offset estimation, degrades the receiver performance. In frequency reuse cellular systems, ACI can be minimized by avoiding the use of adjacent channels within a cell (Chuang et al, 1999). However, as the frequency reuse factor approaches one, the distortions from ACI cannot be neglected in the receiver design.

 

Interference mitigation techniques

The characteristics of CCI and ISI of a wireless communications system is determined by the radio interface and the network topology of the system. Accordingly, a broad range of interference mitigation techniques have been deployed at transmitter and/or receiver as illustrated in figure (1) (Lo et al, 1995).


 

 

 

 

 

 

 

 

 

 

 


Figure 1:  Interference mitigation techniques.

 


In system-design approaches, transmission of co-channel signals is properly managed so that the power of received CCI is maintained below an acceptable level (Laster and Reed, 1997). In contrast, receiver-design approaches actively mitigate the CCI/ISI which cannot be separated by the preemptive system-design approaches. In practical systems, both approaches are employed in joint fashions to reduce the interference (Liang et al, 1996).

 

 

 

Frequency reuse and multiple access

Information streams from multiple users can be transmitted in parallel through a shared radio spectrum by isolating signals from different users in multiple domains. In time, frequency, and code division multiple access (TD/FD/CDMA) techniques, signals from multiple users are transmitted by using non overlapping time slots, non overlapping frequency bands, and codes having very small cross correlations, respectively, so that signals from different users are easily separated. Two forms of CDMA, frequency hopping (FH) and direct sequence (DS), are widely used in military and commercial applications (IEEE, 2003). The second generation digital cellular systems based on IS-95, GSM, and PDC are designed using a combination of the three multiple access techniques to accommodate more channels.

 

Frequency reuse is an example of space division multiple access (SDMA) techniques that separates CCI in cellular systems by utilizing path loss phenomena and radio spectrum partitioning (Laster and Reed, 1997). In a frequency reuse scheme, clustered radio channels are reused in distant co-channel cells in repeating patterns. The transmit power is properly controlled to keep the amount of CCI at a tolerable level. However, the received CIR at a receiver is not guaranteed statistically because of the dynamic nature of the fading channels especially in high capacity wireless systems where more aggressive frequency reuse schemes are employed (Jakes, 1994).

 

Wireless packet networks (WLAN and WPAN) based on IEEE802.11 and Bluetooth standards provide complementary wireless solutions for low-mobility broadband multimedia traffic in the unlicensed ISM band (Li et al, 1999). The WLANs and WPANs operate in two different network topologies: access-point and ad-hoc network. Without any centralized multiple access control among co-located networks, independent multiple access control (MAC) in each network such as carrier sensing multiple access with collision avoidance (CSMA/CA) cannot avoid the collision between packets from different networks. Therefore, CCI from packet collisions can only be mitigated by using direct sequence or frequency hopping spread spectrum techniques at physical (PHY) layer signal processing.

 

Adaptive filtering

Interference cancelling receiver design is often viewed as an adaptive filtering with feedforward and feedback filters as illustrated in Figure 2 (Laster and Reed, 1997). This technique finds its root in adaptive equalization research, which primarily focuses on mitigating ISI with single antenna by using linear and nonlinear techniques. Equalization techniques effectively mitigate CCI as well ISI. Two types of equalizers using linear or nonlinear techniques can be found in many references: symbol-by-symbol equalizers and sequence estimators.

 

The most common structure for the linear equalizer is the transversal filter in which the current and past values of the received signal are weighted by equalizer coefficients and summed to produce the output for symbol-by-symbol decisions on the received symbol sequence. The equalizer coefficients are adjusted to minimize some error criterion. The equalizer that forces ISI to zero is called zero-forcing (ZF) equalizer. The Minimum mean square error (MMSE) equalizer outperforms the ZF equalizer in performance and convergence properties by mitigating the noise enhancement (Duel-Hallen, 1992). Nonlinear decision feedback equalizer (DFE) combined with a linear feed forward filter has been proposed to reduce the effect of noise enhancement from precursor and post cursor ISI. (Lo et al, 1995) showed that a directly adapted Recursive least square (RLS) DFE equalizer outperforms an MMSE equalizer, which employs estimates of channel impulse response and the autocorrelation of interference-plus-noise in frequency selective channels in the presence of CCI . One drawback of the DFE type receivers is error propagation when the desired signal is in a deep fade, or when the received CIR is low. (Uesugi et al, 1996) also proposed a DFE type single/double feedback interference cancelling (SF/DFIC) receiver to mitigate CCI by subtracting the ISI components of the estimated co-channel signals.

 

 

 


 

 

 

 

 

 

 

 

 


Figure 2: An adaptive filter model for interference mitigation.

Table 2: Weight functions of diversity combining techniques with CCI

 

Weight

Notes

EGC

W = [1,….,1]

Co-phased and equally weighted

MRC

W = [9*1d,…….9*Nd] =g*d

ML with CSI

OC

W =αR -1 g*d where R-12I+E[gi*giT]

Optimal in sense of Max. SINR

IRC

Metric = arg min {exp(-gi*R-1giT) }

MLSE from impairment vector

 

 


Spatio-temporal interference mitigation

Faded signal reception results in a large penalty in SNR when the receiver has only one set of received signals from a single antenna. For example, a DFE type receiver with single antenna experiences error propagation during the signal reception in a deep fades. The use of multiple antennas at receiver creates multiple-input multiple-output (MIMO) channels in CCI mitigation (IEEE, 2003) and the existing CCI mitigation techniques for multiple-input single-output (MISO) channels can be extended to spatio-temporal interference mitigation techniques by using diversity combing techniques. One advantage of the spatio-temporal approach is a joint suppression-and-equalization of CCI and ISI (Laster and Reed, 1997).

 

Diversity combining for CCI suppression

Figure 3 illustrates an architecture of the 1 x N diversity combining receiver with channel vectors g = [gld, . . . , gNdj and gj = [gi, . . . , gjr] of the desired and interfering signals, respectively. Weight functions of four different diversity

combining techniques are summarized in Table 2. (Aalo et al, 2000) and (Hafeez et al, 2000) have analyzed the performance of optimum combining (OC) and MRC techniques with non-Gaussian CCI in flat fading channels in terms of outage probability. (Suzuki, 1993) showed that the MRC has an interference cancelling effect in CCI environments. The optimum linear minimum mean square error combining technique for flat fading channels was proposed by (Winters, 1984) by using the channel information of all co-channel signals in updating the antenna weight coefficients. Unlike MRC which mitigates CCI by enhancing the received SINR at the antenna outputs, OC jointly combats the effects of fading and CCI through digital beamforming with a multiple-element spatial diversity combiner (Laster and Reed, 1997). Though OC is not effective in ISI equalization, this drawback can be compensated for by using a concatenated symbol-by-symbol equalizer or sequence estimator.


                       

 

 

 

 

 

 

Figure 3: A schematic of diversity combining.

 


Two-stage interference cancellation

In frequency selective fading channels, all co-channel signals experience ISI. (Liang et al, 1999) worked on two-stage CCI/ISI reduction method was motivated by this observation. In the two-stage interference mitigation, the CCI is suppressed by a space-time filter in the first stage and the ISI is cancelled by a Viterbi type equalizer in the second stage (Qureshi, 1985). (Duel-Hallen, 1992) and (Uesugi et al, 1996) have suggested DFE-based approaches for joint suppression-and-equalization in MIMO channels. (Chuang et al, 1999) also proposed CCI/ISI mitigation for IS-136 TDMA systems by using MMSE spatial-temporal DFE and linear equalizer (LE).

With sequence estimation equalizers, joint MLSE (J-MLSE) is the maximum-likelihood solution to the signal detection in the ISI channels with CCI (Qureshi, 1985).

 

For diversity combining receivers, the received CIRs at antenna branches are generally assumed equal. However, this assumption is not always applicable, especially in diversity combining with directional antennas. (Mallik et al, 2002) showed that the imbalance of Gaussian noise across antenna branches degrades the performance of equal gain combining (EGC) in correlated Rayleigh faded channels, and (Lo et al, 1995) showed that optimal and selective combining receivers with linear/nonlinear equalizers achieve minimum BER when all antenna branches have equal received SNRs.

 

Beamforming and transmit diversity

Beamforming and transmit diversity are two complementary techniques for using multiple antennas in wireless communication systems. Beamforming achieves an array gain by linearly combining tap gains of an antenna array in highly correlated channels while transmit diversity obtains a diversity gain by exploiting the independence among channels (Lee, 1982). In highly correlated line-of-sight (LOS) indoor channels, beamforming techniques provide CCI- mitigation through spatial filtering (Qureshi, 1985). The spatial filtering of CCI is achieved either by shaping beams to have nulls in the directions of co-channel signals or by forming beams to have a large gain in the direction of the desired signal . For this reason, beamforming requires estimation of the direction of arrival (DoA) of the desired or interfering signals.

 

Several variations of beamforming have been proposed: fixed, switched, and adaptive beamforming. In fixed beamforming (FB) networks, an antenna array forms narrow multiple beams in pre-selected directions for low-mobility users and suppresses the interference from outside of the beamwidth. The multichannel multipoint distribution service (MMDS) for broadband wireless access (BWA) is one example of FB networks (Uesugi, 1996). The switched beamforming (SB) technique uses a switch to select the best beam to receive a particular signal in FB networks. As the realizations of the space division multiple access (SDMA) technique, the fixed and switched beamforming have been applied to existing TDMA cellular networks (Suzuki, 1993).

 

Unlike the beamforming technique which changes the radiation pattern of an antenna array to achieve array gains and CCI-mitigation by controlling the weights of array elements in radio frequency (RF) level, the diversity gain of the transmit diversity (TD) is achieved by combining the signals in baseband or intermediate frequency (IF) level (Li et al, 1999). As a result, TD allows a lot of freedom in transmitter/receiver designs by combining coding and space-time diversity techniques. Space-time encoding techniques employed in transmitter helps the separation of transmitted signals at the receiver by using the orthogonality between space-time code matrices. CCI mitigation in transmit diversity is achieved by using the space-time coding as well as the antenna diversity. Another advantage of the transmit diversity scheme is the simplified receiver structure without losing diversity gains.

 

The third generation (3G) wireless communication systems W-CDMA have considered time diversity techniques as their key contributing technologies. Orthogonal TD (OTD) (3GPP, 2005) is an open loop method in which coded interleaved symbols are split into even and odd symbol streams and transmitted using two different Walsh codes. Space-time transmits diversity (STTD) and space-time spreading (STS) (Malik et al, 2002) techniques use Walsh codes and transmit diversity techniques.

Multiuser detection

Distinguished from single-user detection techniques, which treats signals from co-channel users as interference, multi-user detection (MUD) detects all co-channel signals simultaneously. Since MUD techniques not only increase the system capacity but also improve the quality of an individual communication link by eliminating CCI from multi-users (Stuber, 2001), MUD has been an important technology in interference-limited communication systems such as GSM, IS-54/IS-136, and IS-95 regardless of the multiple access schemes (3GPP, 2005). A joint interference cancelling receiver based on DFE technique is suggested for the TDMA cellular system, and MUD receives for TDMA-based GSM, EDGE, and IS-95 systems, respectively, by using joint sequence estimation techniques based on J-MLSE. To reduce the state of the J-MLSE method, reduced-state joint detection algorithms based on the DDFSE technique is been suggested. Also, the single-antenna interference cancellation (SAIC) techniques for TDMA cellular systems is considered practical solution for capacity increase without modifying existing infrastructures. 

 

Conclusion

Fading is traditionally countermeasure by channel coding and interleaving techniques as well as transmit / receive antenna diversity schemes. ISI from multipath reception can be combated by various linear / nonlinear type equalization techniques employing symbol-by-symbol detection methods such as decision feedback equalizer (DFE) or sequence estimation methods such as maximum likelihood sequence estimation (MLSE).

 

In cellular networks, CCI is the interference from neighboring cells using the same radio channels. As the frequency reuse factor decreases to one, CCI is unavoidable due to the channel reuse in adjacent cells. In ad-hoc type wireless networks such as WLAN and WPAN, the signals transmitted from multiple networks operating in close proximity behave as CCI to each other. CCI is best handled by the joint MLSE receiver. Less complex linear filter type receivers suppresses CCI by controlling the filter coefficients in the sense of maximizing the signal-to-interference-plus-noise-ratio.

 

In typical wireless mobile communications systems, CCI, ISI and fading often arise together. Hence, the receiver designs for mitigating these impairments in joint fashions are quite common.

 

References

            Aalo, V. A. and Zhang, J., (2000) “Performance of Antenna Array Systems with Optimum Combining in a Rayleigh Fading Environment,” IEEE Communications., vol. 4, pp. 387-389.

 

            Bottomley, G. E. and Jamal, K., (1995) “Adaptive Arrays and MLSE Equalization,” in Proceedings IEEE Vehicular Technology Conference, VTC ‘95, pp. 50-54.

           

            Bottomley, G. E., Molnnar, K. J., and Chennakeshu, S. (1999) “Interference Cancellation with an Array Processing MLSE Receiver,” IEEE Trans. Vehicular Technology, vol. 48, pp. 1321-1331.

           

           Chuang, J., Cimini, L. J., Li, G. Y., Mcnair, B., Sollenberger, N., Zhao, H., and Suzuki, M., (1999) “High-Speed Wireless Data Access Based on Combining EDGE with Wideband OFDM,” IEEE Communications.,  Magazine, pp. 92-98.

 

            Duel-Hallen, A., (1992) “Equalizers for Multiple Input/Multiple Output Channels and PAM Systems with Cyclostationary Input Sequences,”. IEEE Communications, vol. 10, pp. 630-639.

 

            Haffeez, A., Molnar, K. J., Bottomley G. E., and Ramesh, R. (2000) “Capacity and Quality Enhancement for ANSI-136 Downlink Using Interference Cancellation and Beamforming,” in Proceedings IEEE Vehicular Technology Conference, VTC ‘00, pp. 2414-2421

.

            IEEE, “Low-Rate Wireless PAN Medium Access Control (MAC) and Physical Layer (PHY) Specification,” 802.15.4, 2003

 Jakes, W. C., (1994) Microwave Mobile Communications. New York: IEEE Press.

 

           Laster, J. D. and Reed, J. H. (1997) “Interference Rejection in Digital Wireless Communications,” IEEE Signal Processing Magazine. vol. 14, pp. 37-62.

 

Lee, W.C.Y.(1982): Mobile Communications Engineering. New York: McGraw-Hill.

 

            Liang, J., Paulraj, A. J., and Singh, A. (1996) “Two Stage CCI/ISI Reduction with Space-Time Processing in TDMA Cellular Networks,” in Proceedings IEEE Thirty Asilomar Conference, pp. 607-611.

 

             Li, Y., Winters, J. H., and Sollenberger, N. (1999) “Spatial-Temporal Equalization for IS-136 Systems with Rapid Dispersive Fading and Co-Channel Interference,” IEEE Trans. Vehicular. Technol., vol. 48, pp. 1182-1194.

 

            Lo, N. W. K., Falconer, D., and Sheikh, A. U. H. (1995) “Adaptive Equalization for Co-Channel Interference in a Multipath Fading Environment,” IEEE Trans. Communications., vol. 43, pp. 1441-1453.

 

            Mallik, R. K., Win, M. Z., and Winters, J. H. (2002) “Performance of Dual-Diversity Prediction  EGC in Correlated Rayleigh Fading with Unequal Branch SNRs,” IEEE Trans. Communications., vol. 50, pp. 1041-1044.        

 

            Qureshi, S. (1985) “Adaptive Equalization,” in Proceedings of IEEE, vol. 73, pp. 1349-1387

.

            Stuber, G. L. (2001): Principles of Mobile Communications, 2nd ed. Boston: Kiuwer Academic Publishers. 

 

Suzuki, H. (1993): “Signal Transmission Characteristics of Diversity Reception with Least-Squares Combining-Relationship between Desired Signal Combining and InterferenceCancelling,” Electronics and communications magazine, Japan, Part I, vol. 76, pp. 80-93.

 

            Uesugi, M., Futagi, S., and Homma, K. (1996): “Interference Cancellation Method using DFE,” in Proceedings IEEE Vehicular Technology Conference, VTC ‘96, pp. 1190-1194.

 

            Winters, J. H. (1984) “Optimum Combining in Digital Mobile Radio with Co-channel Interference,” IEEE J. Select. Areas Communications., vol. 2, pp. 528-539. 

 

            3GPP, “3GPP TSG GERAN Radio Transmission and Reception,” 3GPP TS 45, 2005, Release 5.