【摘要】：基于Jacobi迭代提出一種低復(fù)雜度信號檢測算法,在算法實(shí)現(xiàn)中避免了矩陣求逆運(yùn)算.數(shù)學(xué)推導(dǎo)證明,該算法應(yīng)用于MMSE檢測時是收斂的,與傳統(tǒng)的Neumann級數(shù)展開方法對比,能達(dá)到與其完全相同的檢測性能,并且在任意迭代次數(shù)下能將復(fù)雜度保持在O(K~2),而后者當(dāng)級數(shù)展開項(xiàng)數(shù)大于等于3時復(fù)雜度上升為O(K~3).為了進(jìn)一步將Jacobi迭代應(yīng)用到軟判決中,提出了一種用于信道譯碼的LLR的近似計算方法.仿真結(jié)果表明,經(jīng)過幾次迭代,Jacobi迭代算法收斂較快,并接近MMSE檢測性能.
[Abstract]:Based on Jacobi iteration, a low complexity signal detection algorithm is proposed, which avoids the inverse operation of matrix in the implementation of the algorithm. The mathematical derivation proves that the algorithm is convergent when applied to MMSE detection. Compared with the traditional Neumann series expansion method, the algorithm can achieve exactly the same detection performance, and can keep the complexity at O (K 鈮，

本文編號：2510901