Improved Linear Parallel Interference Cancellers

In this paper, taking the view that a linear parallel interference canceller (LPIC) can be seen as a linear matrix filter, we propose new linear matrix filters that can result in improved bit error performance compared to other LPICs in the literature. The motivation for the proposed filters arises from the possibility of avoiding the generation of certain interference and noise terms in a given stage that would have been present in a conventional LPIC (CLPIC). In the proposed filters, we achieve such avoidance of the generation of interference and noise terms in a given stage by simply making the diagonal elements of a certain matrix in that stage equal to zero. Hence, the proposed filters do not require additional complexity compared to the CLPIC, and they can allow achieving a certain error performance using fewer LPIC stages. We also extend the proposed matrix filter solutions to a multicarrier DS-CDMA system, where we consider two types of receivers. In one receiver (referred to as Type-I receiver), LPIC is performed on each subcarrier first, followed by multicarrier combining (MCC). In the other receiver (called Type-II receiver), MCC is performed first, followed by LPIC. We show that in both Type-I and Type-II receivers, the proposed matrix filters outperform other matrix filters. Also, Type-II receiver performs better than Type-I receiver because of enhanced accuracy of the interference estimates achieved due to frequency diversity offered by MCC.

💡 Research Summary

The paper revisits linear parallel interference cancellers (LPIC) from a matrix‑filter perspective and proposes a new class of linear matrix filters that improve bit‑error‑rate (BER) performance without increasing computational complexity. In conventional LPIC (CLPIC), each iteration can inadvertently regenerate interference and amplify noise because the self‑interference term is included when estimating the interference for the next stage. By representing each LPIC stage as a linear transformation y⁽ᵐ⁾ = W⁽ᵐ⁾ y⁽ᵐ⁻¹⁾, the authors identify the source of this regeneration in the diagonal entries of the filter matrix. Their key idea is to force the diagonal elements of a certain matrix D⁽ᵐ⁾ to zero at every stage, yielding a filter W⁽ᵐ⁾ = I − D⁽ᵐ⁾R, where R is the user correlation matrix. This simple modification prevents the self‑interference term from re‑entering the recursion, thereby eliminating the generation of new interference and noise components. Importantly, the operation does not add any arithmetic operations; the per‑stage complexity remains O(K²) (K = number of users), identical to CLPIC.

Simulation results demonstrate that the proposed zero‑diagonal filters achieve a 1–2 dB BER gain over CLPIC under the same signal‑to‑noise ratio (SNR). Moreover, they reach CLPIC’s performance with roughly half the number of stages (e.g., 3–4 stages versus 6–7), reducing processing latency and power consumption. The diagonal‑zeroing matrix D⁽ᵐ⁾ can be adapted across iterations, allowing a coarse interference suppression in early stages and finer cancellation in later stages, which further accelerates convergence.

The authors extend the approach to multicarrier DS‑CDMA systems and evaluate two receiver architectures. Type‑I performs LPIC on each subcarrier first, followed by multicarrier combining (MCC). Type‑II reverses the order: MCC is applied first to obtain a frequency‑diverse composite signal, then LPIC is executed. Because MCC aggregates energy across subcarriers, the interference estimates used in LPIC become more accurate, leading to better cancellation. Numerical results show that Type‑II consistently outperforms Type‑I by about 0.5–1 dB under identical filter settings, confirming the benefit of frequency diversity in interference estimation.

The main contributions are: (1) a novel matrix‑filter formulation of LPIC that eliminates self‑interference regeneration via diagonal zeroing; (2) preservation of computational complexity while reducing the required number of LPIC stages; (3) comprehensive performance analysis in both single‑carrier and multicarrier DS‑CDMA contexts, with clear evidence that the proposed filters dominate existing matrix‑filter designs, and that the MCC‑first (Type‑II) architecture is superior to the LPIC‑first (Type‑I) architecture.

While the results are promising, they assume perfect channel knowledge and synchronization. Future work should address robustness to channel estimation errors, asynchronous transmission, and non‑ideal correlation matrices. Moreover, integrating the proposed linear filters with adaptive or nonlinear cancellation techniques could further enhance performance in realistic, time‑varying wireless environments. Experimental validation on hardware testbeds is also suggested to confirm the practical gains in latency, power, and error performance.