# US 2010/0167679 A1 · Method and Arrangement for Improved Model Order Selection

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2010/0167679 A1
- Document ID: US20100167679A1
- Status: Published application
- Kind: A1
- Application number: 12/601,568
- Filed: 2007-05-28
- Published: 2010-07-01
- Earliest claimed: 2007-05-28
- Assignees: Telefonaktiebolaget LM Ericsson (publ)
- Inventors: Miguel Lopez; Gustav Almquist; Tomas Andersson; Dennis Hui
- Web page: https://patentgrep.com/patent/US20100167679A1

## CPC

- H04L 25/0216: {with estimation of channel length}
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

In a method of enabling model order selection for joint channel synchronization and noise covariance estimation of at least one received signal in a wireless communication network, generating S0 a spatially and temporally stacked signal model by stacking successive samples of temporally adjacent received signal vectors and corresponding training vectors, computing S1 a noise variance matrix for each hypothesized synchronization position, channel length and stacking order, based on the stacked training symbols: determining S2 a best synchronization position for the received signal, based on the stacked training vectors, by jointly determining the best synchronization position for the received signal and estimating a channel length and a stacking order for said signal model based on the stacked training vectors.

## Claims

### Claim at position 1

1-11. (canceled)

### Claim 12

A method of enabling model order selection for joint channel synchronization and noise covariance estimation of at least one received signal in a wireless communication network, said method comprising the steps of:
generating a spatially and temporally stacked signal model by stacking successive samples of temporally adjacent received signal vectors and corresponding training vectors;
computing a noise variance matrix for each hypothesized synchronization position, channel length and stacking order, based on the stacked training symbols; and
determining a best synchronization position for the received signal, based on the stacked training vectors;
wherein said determining step comprises jointly determining the best synchronization position for the received signal and estimating a channel length and a stacking order for said signal model based on the stacked training vectors.

### Claim 13 (depends on 12)

The method of claim 12, wherein said determining step comprises:
defining a model parameter dependent on a stacking order of said generated stacked signal model; and
calculating an estimate of the channel length, the stacking order and the best synchronization position based on a minimum of a function dependent of the determinant of said noise covariance matrix, said model parameter and a penalty function.

### Claim 14 (depends on 13)

The method of claim 13, wherein said determining step comprises the further steps of:
calculating the estimate of the channel length, the stacking order, and the best synchronization position based on a minimum of a function dependent of the determinant of said noise covariance matrix to the power of said model parameter and a predetermined exponential function dependent on a penalty function.

### Claim 15 (depends on 14)

The method of claim 14, further comprising calculating said estimate of the channel length, the stacking order and the best synchronization position based on a minimum of any monotonically increasing mapping of said function.

### Claim 16 (depends on 14)

The method of claim 14, wherein said penalty function p(L,M) is a positive function of L and M, where L is the channel length, and M is the stacking order.

### Claim 17 (depends on 16)

The method of claim 16, wherein said penalty function p(L,M) is represented by the expression
p(L,M)=CL+DM,
where C, D are positive constants, L is the channel length, and M is the stacking order.

### Claim 18 (depends on 17)

The method of claim 17, characterized in that C=2 and D=2.

### Claim 19 (depends on 13)

The method of claim 13, wherein said model parameter γ(M) is represented by the expression
γ(M)=1/((M+1)Nr)
where M is the stacking order, and Nr is the number of branches in the received signal.

### Claim 20

A receiver arrangement configured for enabling model order selection for joint channel synchronization and noise covariance estimation of at least one received signal in a wireless communication network, said receiver arrangement comprising:
a stacking circuit configured for generating a spatially and temporally stacked signal model by stacking successive samples of temporally adjacent received signal vectors and corresponding training vectors;
a noise variance estimation circuit configured for computing a noise variance matrices for all hypothesized synchronization positions, channel lengths and stacking orders, based on the stacked training symbols;
a synchronization determining circuit configured for determining a best synchronization position for the received signal, based on the stacked training vectors; and
wherein said synchronization determining circuit is configured for jointly determining the best synchronization position for the received signal and estimating a channel length and a stacking order for said signal model based on the stacked training vectors.

### Claim 21 (depends on 20)

The receiver arrangement of claim 20, wherein said receiver arrangement comprises a mobile station receiver arrangement implemented in a mobile station configured for operation in the wireless communication network.

### Claim 22 (depends on 20)

The receiver arrangement of claim 20, wherein said receiver arrangement comprises a base station receiver arrangement implemented in a base station configured for operation in the wireless communication network.