随机过程基础(5)---各态历经性(2)、联合分布、连续信号功率谱密度(PSD)

各态历经性

可以证明,随机过程X(t)具有均值遍历性的充要条件是 lim ⁡ Y → ∞ 1 T ∫ 0 2 T ( 1 − τ 2 T ) [ R X ( τ ) − m X 2 ] d τ = 0 \lim_{Y\rightarrow \infty}\frac{1}{T}\int_0^{2T}(1-\frac{\tau}{2T})[R_X(\tau)-m_X^2]d\tau =0 YlimT102T(12Tτ)[RX(τ)mX2]dτ=0具有相关函数遍历性的充要条件是 lim ⁡ Y → ∞ 1 T ∫ 0 2 T ( 1 − τ 2 T ) [ R ϕ ( τ ) − R X 2 ( τ ) ] d τ = 0 \lim_{Y\rightarrow \infty}\frac{1}{T}\int_0^{2T}(1-\frac{\tau}{2T})[R_\phi(\tau)-R_X^2(\tau)]d\tau =0 YlimT102T(12Tτ)[Rϕ(τ)RX2(τ)]dτ=0对于零均值的平稳正态随机信号,如果RX(τ)连续,则各态历经性的充要条件简化为 ∫ 0 ∞ ∣ R X ( τ ) ∣ d τ < ∞ \int_0^\infty|R_X(\tau)| d\tau<\infty 0RX(τ)dτ<
例如,判断随机相位信号的各态历经性:
设有随机相位信号 X ( t ) = A c o s ( ω 0 t + ϕ ) X(t)=Acos(\omega_0t+\phi) X(t)=Acos(ω0t+ϕ)则X(t)的均值为0,自相关函数 R X ( t + τ , t ) = E [ X ( t + τ ) X ( t ) ] R_X(t+\tau,t)=E[X(t+\tau)X(t)] RX(t+τ,t)=E[X(t+τ)X(t)] = A 2 E { c o s [ ω 0 ( t + τ ) + ϕ ] c o s ( ω 0 t + ϕ ) } =A^2E\{cos[\omega_0(t+\tau)+\phi]cos(\omega_0t+\phi)\} =A2E{ cos[ω0(t+τ)+ϕ]cos(ω0t+ϕ)}应用积化和差公式, = A 2 2 { E [ c o s ( 2 ω 0 t + ω o τ + 2 ϕ ) ] + E [ c o s ω 0 τ ] } =\frac{A^2}{2}\{E[cos(2\omega_0t+\omega_o\tau+2\phi)]+E[cos\omega_0\tau]\} =2A2{ E[cos(2ω0t+ωoτ+2ϕ)]+E[cosω0τ]} = A 2 2 c o s ω 0 τ =\frac{A^2}{2}cos\omega_0\tau =2A2cosω0τ又时间平均 m X ‾ = 1 ⋅ i ⋅ m 1 2 T ∫ − T T A c o s ( ω 0 + ϕ ) d t = 0 \overline{m_{X}}=1\cdot i\cdot m\frac{1}{2T}\int_{-T}^{T}Acos(\omega_0+\phi)dt=0 mX=1im2T1TTAcos(ω0+ϕ)dt=0时间相关函数 R X ( τ ) ‾ = 1 ⋅ i T → ∞ ⋅ m 1 2 T ∫ − T T X ( t + τ ) X ( t ) d t \overline{R_X(\tau)}=1\cdot i_{T\rightarrow\infty}\cdot m\frac{1}{2T}\int_{-T}^{T}X(t+\tau)X(t)dt RX(τ)=1iTm2T1TTX(t+τ)X(t)dt = 1 ⋅ i T → ∞ ⋅ m 1 2 T ∫ − T T A c o s [ ω 0 ( t + τ ) + ϕ ] A c o s ( ω 0 t + ϕ ) d t =1\cdot i_{T\rightarrow\infty}\cdot m\frac{1}{2T}\int_{-T}^{T}Acos[\omega_0(t+\tau)+\phi]Acos(\omega_0t+\phi )dt =1iTm2T1TTAcos[ω0(t+τ)+ϕ]Acos(ω0t+ϕ)dt = A 2 2 1 ⋅ i T → ∞ ⋅ m 1 2 T ∫ − T T [ c o s ( 2 ω 0 t + ω 0 τ + 2 ϕ ) + c o s ( ω 0 τ ] d t =\frac{A^2}{2}1\cdot i_{T\rightarrow\infty}\cdot m\frac{1}{2T}\int_{-T}^{T}[cos(2\omega_0t+\omega_0\tau+2\phi)+cos(\omega_0\tau] dt =2A21iTm2T1TT[cos(2ω0t+ω0τ+2ϕ)+cos(ω0τ]dt = A 2 2 c o s ω 0 τ =\frac{A^2}{2}cos\omega_0\tau =2A2cosω0τ故时间量和统计量两两对应相等,可见是各态历经过程。
对于大多数随机过程,都具有各态历经性。工程上一般直接认为平稳随机信号是各态历经的。设x(t)是一条样本函数,就用 m X ^ = 1 2 T ∫ − T T x ( t ) d t \hat{m_X}=\frac{1}{2T}\int_{-T}^T x(t)dt mX^=2T1TTx(t)dt估计均值,用 R ^ X ( τ ) = 1 2 T ∫ − T T x ( t + τ ) x ( t ) d t \hat{R}_X(\tau)=\frac{1}{2T}\int_{-T}^T x(t+\tau)x(t)dt R^X(τ)=2T1TTx(t+τ)x(t)dt估计自相关函数,得到的结果差距不大。
对于随机序列X(n),对应地, m X ^ = 1 N ∑ N = 0 N − 1 x ( n ) \hat{m_X}=\frac{1}{N}\sum_{N=0}^{N-1}x(n) mX^=N1N=0N1x(n) σ X 2 ^ = 1 N − 1 ∑ n = 0 N − 1 [ x ( n ) − m ^ X ] 2 \hat{\sigma_X^2}=\frac{1}{N-1}\sum_{n=0}^{N-1}[x(n)-\hat m_X]^2 σX2^=N11n=0N1[x(n)m^X]2 R X ( m ) ^ = 1 N − ∣ m ∣ ∑ n = 0 N − ∣ m ∣ − 1 x ( n ) x ( n + m ) \hat{R_X(m)}=\frac{1}{N-|m|}\sum_{n=0}^{N-|m|-1}x(n)x(n+m) RX(m)^=Nm1n=0

评论 1
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值