Wireless Information Transmission System Lab. Interference 2006/3/9 王森弘 Institute of Communications Engineering National Sun Yat-sen University
Introduction Interference Outline Multiuser Interference (MUI) in the MC-CDMA System. Inter-Carrier Interference (ICI) in the OFDM System. Multiuser Interference (MUI) in the OFDMA System. Interference Cancellation Techniques Parallel Interference Cancellation (PIC) Successive Interference Cancellation (SIC) 2
Introduction The multiuser systems have the interesting property that their capacity is typically limited by multiple-access interference (MAI) or multiuser interference (MUI), rather than noise. Two quite different, suboptimal approaches for multiuser detection (MUD) emerged, which had much lower complexity than the optimum multiuser detector: interference cancellation (IC). adaptive filtering. The IC techniques can broadly broken into successive (serial) and parallel schemes for canceling MAI. 3
MC-CDMA In the MC-CDMA system, the multiuser interference due to the channel effect which cause the orthogonality between different codes loss. For example, [1 1 1 1] and [1-1 1-1] are two different orthogonal code. These code will spread on different subcarriers. When the channel responses are [0.8 0.5 0.6 0.7], in the receiver, these two codes will become [0.8 0.5 0.6 0.7] and [0.8-0.5 0.6-0.7]. The correlation between these codes will become 0.64-0.25+0.36-0.49=0.26. 4
Inter-Carrier Interference (ICI) In the OFDM system, when the orthogonality between different subcarriers loss, it cause inter-carrier interference. X 0 Modulation Serial to Parallel Converter X 1 IFFT Parallel to Serial Converter Add Guard Interval x n X N 1 2 exp j πε n N z n Demodulation Serial to Parallel Converter FFT Parallel to Serial Converter Remove Guard Interval y n 5
Inter-Carrier Interference (ICI) After inverse discrete Fourier transform (IDFT) and guard-interval addition, the time-domain discrete signal is given by N 1 j 2π ni 1 N xn = Xie, N n N 1 g N i= 0 At the receiver, the received baseband signal, which is corrupted by the additive white Gaussian noise (AWGN), is given by j2πε n N n = n + n y x e z 6
Inter-Carrier Interference (ICI) After the guard-interval removal and DFT processing of the received signal, the received signal on the mth subcarrier is expressed a Y m N 1 1 j2 ynexp N n= 0 = π nm N ( + ε) N 1 N 1 1 j2πn i m = X i exp N i= 0 n= 0 N ( + ε ) j2π N i m 1 exp N 1 1 N = X i N i= 0 j2π ( i m+ ε) 1 exp N Neglect noise Geometric series 7
Inter-Carrier Interference (ICI) { exp( jπ ( i m+ ε) ) exp( jπ ( i m+ ε) )} exp( jπ ( i m+ ε) ) π ( + ε) π ( + ε) π ( + ε) N 1 1 = X i N i= 0 j i m j i m j i m exp exp exp N N N ( π ( ε) ) π ( i m+ ε) N 1 1 sin i m+ N 1 = Xi exp j ( i m ε ) N + i= 0 N sin N N 1 1 sin ( πε ) N 1 sin ( π ( i m+ ε) ) N 1 = Xm exp j ε Xi exp j ( i m ε) N πε N + + i= 0 π sin ( i m ε) N + i m sin N N Desired carrier ICI terms 8
OFDMA In the OFDMA system, there is no multiuser interference in the downlink. Because the orthogonality between different subcarriers can be restored by carrier frequency offset (CFO) estimation and carrier frequency offset compensation. It is the same as the OFDM system. In the uplink, the basestaion (BS) and the user terminals (UT) have different oscillators. It cause the CFOs between different UTs and BS are different. In the BS, it can not use CFO estimation and CFO compensation to restore the orthogonality between different subcarriers. 9
OFDMA In the OFDMA system, it similar to the OFDM system. The CFOs will destroy the orthogonality between the subcarriers. Therefore, the inter-carrier interference terms which interfere with other UT s subcarriers become the multiuser interference. 10
OFDMA System User 1's Data 1 X i Subcarriers Placement IDFT Add Guard Interval x 1 n DAC RF 1 h n User K's Data K X i Subcarriers Placement IDFT Add Guard Interval K x n DAC RF K h n RF ADC Remove Guard Interval r 11
Multiuser Interference in OFDMA After inverse discrete Fourier transform (IDFT) and guard-interval addition, the time-domain discrete signal of the kth user is given by x j 2π ni 1 = X e i, N n N 1 g N i Γ k k N n k At the BS, the total received baseband signal, which is corrupted by the additive white Gaussian noise (AWGN), from all the users is given by K j2πεk n k N n = k k k n + n yn = xn hn k= 1 r y e z 12
Multiuser Interference in OFDMA After the guard-interval removal and DFT processing of the total received signal, the received signal on the ith subcarrier is expressed as R x h e w K j2πε ( ( ) ( ) FFT ) kn k k N i = N n n + n k = 1 π = + + N N 1 K 1 ( ( k) ( k) ) 2 q Xm H exp ( ) k mk j mk i ε k Wi q= 0 k= 1 m Γ N k k ( ) ( ) sin π ( k + ε k k) K 1 m i N 1 = Z exp j ( m i ε ) W k N k 1 m π + + = N k Γ k sin ( mk i ε + k) N ( k) ( k) ( k) Z = X H m m m k k k m k k i 13
Let K () l Multiuser Interference in OFDMA ( k ) ( ε ) ( k ) ( ε ) R = Z I m i+ + W i mk k k i k= 1 m Γ k k= 1 mk Γ k l k k k l l () l ( ε ) ( ε ) = Z I + Z I m i+ i l ml l l m Γl m i K sin ( πv) 1 N 1 I( v) = exp j v. N π N sin v N Desired carrier + Z I m i+ + W m k k i MUI terms i Γ l ICI terms 14
Parallel Interference Cancellation The parallel interference cancellation technique processes all users at the same time. It will detect all users information bits, regenerate all users interference, and subtract all users interference from the composite signal. Finally, it will detect all users information bits again. 15
Parallel Interference Cancellation User 1's First Detection User 1's Regeneration User 1's Cancellation User 1's Cancellation Signal User 1's Second Detection User 1's Transmitted Data Received signal User 2's First Detection User 2's Regeneration User 2's Cancellation User 2's Cancellation Signal User 2's Second Detection User 2's Transmitted Data User K's First Detection User K's Regeneration User K's Cancellation User K's Cancellation Signal User K's Second Detection User K's Transmitted Data 16
Parallel Interference Cancellation First Detection The first detection stage used to quasi-detect the user s transmitted bits. Regeneration In the regeneration stage, according to the user s bits which obtained in the detection stage, it will regenerate the user s multiuser interference terms. Cancellation In cancellation stage, it will subtract other users multiuser interference terms, which generated in the regeneration stage, from the received composite signal. Second Detection The second detection stage used to precise detect the user s transmitted bits. 17
Successive Interference Cancellation The successive interference cancellation technique processes one user at a time. It will detect the user s information bits, regenerate the user s interference, and subtract the interference cause by this user from the composite signal. Then, it will process the next user until all the users processed. 18
Successive Interference Cancellation Received signal Cancellation User's Cancellation Signal Detection User's Transmitted Data Regeneration 19
Successive Interference Cancellation Cancellation Initially, let the received composite signal be the first user s cancellation signal. Otherwise, it will subtract multiuser interference term, which generated in the regeneration stage, from the received composite signal successively. Detection The detection stage used to detect the user s transmitted bits. Regeneration In the regeneration stage, according to the user s bits which obtained in the detection stage, it will regenerate the user s multiuser interference terms. 20
Successive Interference Cancellation 21
Inter-Carrier Interference Reference J. Armstrong, Analysis of new and existing methods of reducing intercarrier interference due to carrier frequency offset in OFDM, IEEE Trans. Commun., vol. 47, no. 3, pp. 365-369, March 1999. Y. Zhao, and S. Haggman, Intercarrier interference self-cancellation scheme for OFDM mobile communication systems, IEEE Trans. Commun., vol. 49, no. 7, pp. 1185-1191, July 2001. Kusha Panta and Jean Armstrong, Spectral Analysis of OFDM Signals and its Improvement by Polynomial Cancellation Coding," IEEE Tran. Consumer Electronics, Volume: 49, Issue: 4, pp.939 943, Nov. 2003. Katherine A. Seaton and Jean Armstrong, Polynomial Cancellation Coding and Finite Differences," IEEE Trans. Information Theory, Vol. 46, No. 1, pp311-313, Jan. 2000. 22
Parallel Interference Cancellation Reference Yukitoshi Sanada and Masao Nakagawa, A multiuser interference cancellation technique utilizing convolutional codes and orthogonal multicarrier modulation for wireless indoor communications, Selected Areas in Communications, IEEE Journal on, Vol. 14, No. 8, pp. 1500-1509, Oct. 1996. Defeng (David) Huang and Khaled Ben Letaief, An Interference-Cancellation Scheme for Carrier frequency Offsets Correction in OFDMA Systems, IEEE Trans. on Comm., Vol. 53, No. 7, pp. 1155-1165, July 2005. Successive Interference Cancellation Jeffrey G. Andrews and Teresa H. Y. Meng, Performance of Multicarrier CDMA With Successive Interference Cancellation in a Multipath Fading Channel, IEEE Trans. Commun., Vol. 52, No. 5, pp.811-822, May 2004.. 23