AD9361相关

AD9361示意图

典型特性

接收通路

TIA_LPF和BB_LPF属于模拟滤波器，HB3/DEC3、HB2、HB1、FIR属于数字滤波器。

发射通路

RF AND BBPLL SYNTHESIZER

经由片上的DCXO或外部晶振，再经过缓冲、2倍频、2分频、4分频之后，能够得到三个独立的Fref，分别作为BBPLL、Tx、Rx的参考频率。为了RFPLL能够有最好的性能，ADI建议将参考时钟或晶振频率在scale之后设置的越接近80MHz越好。

RF PLL

AD9361片内集成了整个PLL（包括VCO和环路滤波器），PLL运行在6~12GHz。

PLL经过divider之后，获得70MHz~6GHz的射频频率。

BBPLL

BB PLL是用来产生AD9361片内的数字时钟，包括ADC CLK、DAC CLK、所有的模拟校正时钟等，PLL运行在715MHz~1.43GHz。

REF_CLK、BBPLL_CLK、RFPLL_CLK的频率范围

摘自《AD9361_cn》

摘自《AD9361 Reference Manual》

摘自《AD9361配置软件》

ENABLE STATE MACHINE GUIDE

TDD 同一时刻只能发送或接收
FDD 可以收发同时
使能状态机(ENSM)能实时控制器件的当前状态。

Tx Power Control

发射功率的控制是通过调节发射衰减实现的，AD9361片上集成了一个hard-coded的衰减查找表，查找表的深度为360，衰减步进为0.25dB，总共可调衰减为(360-1)*0.25=89.75dB。

2R2T的收发时序

摘自《AD9361 Reference Manual》

FIR系数

SPI配置

复位期间不能进行SPI操作

复位引脚是低有效的
当一直复位时，似乎不能进行SPI操作，如SPI回读

默认先传高比特

MSB传输时，当启用多字节传输时，寄存器地址是递减的

SPI_CLK的最大频率（时钟最大50M）

官方范例代码分析

收发链路上的时钟

摘自《AD9361配置软件》

上图中的内容，可以在官方示例代码中找到对应内容，如下所示，

//摘自main.c
AD9361_InitParam default_init_param = {
    //以上省略
    
	/* Reference Clock */
	40000000UL,	//reference_clk_rate
    
    //以上省略
    /* Rate & BW Control */
	{983040000, 245760000, 122880000, 61440000, 30720000, 30720000},// rx_path_clock_frequencies[6] *** adi,rx-path-clock-frequencies
	{983040000, 122880000, 122880000, 61440000, 30720000, 30720000},// tx_path_clock_frequencies[6] *** adi,tx-path-clock-frequencies
	18000000,//rf_rx_bandwidth_hz *** adi,rf-rx-bandwidth-hz
	18000000,//rf_tx_bandwidth_hz *** adi,rf-tx-bandwidth-hz
    
    //以下省略
}

//摘自ad9361.h
enum ad9361_clocks {
	BB_REFCLK,
	RX_REFCLK,
	TX_REFCLK,
	BBPLL_CLK,
	ADC_CLK,
	R2_CLK,
	R1_CLK,
	CLKRF_CLK,
	RX_SAMPL_CLK,
	DAC_CLK,
	T2_CLK,
	T1_CLK,
	CLKTF_CLK,
	TX_SAMPL_CLK,
	RX_RFPLL_INT,
	TX_RFPLL_INT,
	RX_RFPLL_DUMMY,
	TX_RFPLL_DUMMY,
	RX_RFPLL,
	TX_RFPLL,
	NUM_AD9361_CLKS,
	EXT_REF_CLK,
};

//用于对adi,rx-path-clock-frequencies进行索引
enum ad9361_pdata_rx_freq {
	BBPLL_FREQ,
	ADC_FREQ,
	R2_FREQ,
	R1_FREQ,
	CLKRF_FREQ,
	RX_SAMPL_FREQ,
	NUM_RX_CLOCKS,
};

//摘自ad9361.c
int32_t ad9361_init (struct ad9361_rf_phy **ad9361_phy, AD9361_InitParam *init_param)
{
	//以上省略
	
	/* Reference Clock */
	phy->clk_refin->rate = init_param->reference_clk_rate;
     
    //以上省略
    /* Rate & BW Control */
	for(i = 0; i < 6; i++) {
		phy->pdata->rx_path_clks[i] = init_param->rx_path_clock_frequencies[i];
	}
	for(i = 0; i < 6; i++) {
		phy->pdata->tx_path_clks[i] = init_param->tx_path_clock_frequencies[i];
	}
	phy->pdata->rf_rx_bandwidth_Hz = init_param->rf_rx_bandwidth_hz;
	phy->pdata->rf_tx_bandwidth_Hz = init_param->rf_tx_bandwidth_hz;	

	//以下省略
}

//摘自ad9361.c
/**
 * Set the RX and TX path rates.
 * @param phy The AD9361 state structure.
 * @param rx_path_clks RX path rates buffer.
 * @param tx_path_clks TX path rates buffer.
 * @return 0 in case of success, negative error code otherwise.
 */
int32_t ad9361_set_trx_clock_chain(struct ad9361_rf_phy *phy,uint32_t *rx_path_clks,uint32_t *tx_path_clks)
{
	//以上省略
    ret = ad9361_validate_trx_clock_chain(phy, rx_path_clks);
	if (ret < 0)    return ret;

	ret = clk_set_rate(phy, phy->ref_clk_scale[BBPLL_CLK], rx_path_clks[BBPLL_FREQ]);
	if (ret < 0)    return ret;

	for (i = ADC_CLK, j = DAC_CLK, n = ADC_FREQ;
		i <= RX_SAMPL_CLK; i++, j++, n++) {
		ret = clk_set_rate(phy, phy->ref_clk_scale[i], rx_path_clks[n]);
		if (ret < 0) {
			dev_err(dev, "Failed to set BB ref clock rate (%"PRId32")",
				ret);
			return ret;
		}
		ret = clk_set_rate(phy, phy->ref_clk_scale[j], tx_path_clks[n]);
		if (ret < 0) {
			dev_err(dev, "Failed to set BB ref clock rate (%"PRId32")",
				ret);
			return ret;
		}
	}
}

/***************************************************************************//**
 * @brief clk_set_rate
*******************************************************************************/
int32_t clk_set_rate(struct ad9361_rf_phy *phy,
					 struct refclk_scale *clk_priv,
					 uint32_t rate)
{
    //以上省略
	source = clk_priv->source;
	if(phy->clks[source]->rate != rate)
	{
		switch (source) {
			case TX_REFCLK:
			case RX_REFCLK:
			case BB_REFCLK:
				round_rate = ad9361_clk_factor_round_rate(clk_priv, rate,
								&phy->clk_refin->rate);
				ad9361_clk_factor_set_rate(clk_priv, round_rate,
						phy->clk_refin->rate);
				phy->clks[source]->rate = ad9361_clk_factor_recalc_rate(clk_priv,
												phy->clk_refin->rate);
				break;
			case TX_RFPLL_INT:
			case RX_RFPLL_INT:
				round_rate = ad9361_rfpll_int_round_rate(clk_priv, rate,
								&phy->clks[clk_priv->parent_source]->rate);
				ad9361_rfpll_int_set_rate(clk_priv, round_rate,
						phy->clks[clk_priv->parent_source]->rate);
				phy->clks[source]->rate = ad9361_rfpll_int_recalc_rate(clk_priv,
											phy->clks[clk_priv->parent_source]->rate);
				break;
			case RX_RFPLL_DUMMY:
			case TX_RFPLL_DUMMY:
				ad9361_rfpll_dummy_set_rate(clk_priv, rate);
				break;
			case TX_RFPLL:
			case RX_RFPLL:
				round_rate = ad9361_rfpll_round_rate(clk_priv, rate);
				ad9361_rfpll_set_rate(clk_priv, round_rate);
				phy->clks[source]->rate = ad9361_rfpll_recalc_rate(clk_priv);
				break;
			case BBPLL_CLK:
				round_rate = ad9361_bbpll_round_rate(clk_priv, rate,
								&phy->clks[clk_priv->parent_source]->rate);
				ad9361_bbpll_set_rate(clk_priv, round_rate,
					phy->clks[clk_priv->parent_source]->rate);
				phy->clks[source]->rate = ad9361_bbpll_recalc_rate(clk_priv,
											phy->clks[clk_priv->parent_source]->rate);
				phy->bbpll_initialized = true;
				break;
			case ADC_CLK:
			case R2_CLK:
			case R1_CLK:
			case CLKRF_CLK:
			case RX_SAMPL_CLK:
			case DAC_CLK:
			case T2_CLK:
			case T1_CLK:
			case CLKTF_CLK:
			case TX_SAMPL_CLK:
				round_rate = ad9361_clk_factor_round_rate(clk_priv, rate,
								&phy->clks[clk_priv->parent_source]->rate);
				ad9361_clk_factor_set_rate(clk_priv, round_rate,
						phy->clks[clk_priv->parent_source]->rate);
				phy->clks[source]->rate = ad9361_clk_factor_recalc_rate(clk_priv,
											phy->clks[clk_priv->parent_source]->rate);
				break;
			default:
				break;
		}
		//以下省略
    }
    //以下省略
	return 0;
}

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186

收发链路上的时钟处理相关函数的调用关系

AGC

#define REG_AGC_LOCK_LEVEL			     0x101 /* AGC Lock Level */
#define REG_ADC_NOISE_CORRECTION_FACTOR	 0x102 /* ADC noise Correction Factor */
#define REG_GAIN_STP_CONFIG1			 0x103 /* Gain Step Config1 */
#define REG_ADC_SMALL_OVERLOAD_THRESH    0x104 /* ADC Small Overload Threshold */
#define REG_ADC_LARGE_OVERLOAD_THRESH	 0x105 /* ADC Large Overload Threshold */

#define REG_GAIN_TABLE_ADDRESS			 0x130 /* Gain Table Address */
#define REG_GAIN_TABLE_WRITE_DATA1		 0x131 /* Gain Table Write Data1 */
#define REG_GAIN_TABLE_WRITE_DATA2		 0x132 /* Gain Table Write Data2 */
#define REG_GAIN_TABLE_WRITE_DATA3		 0x133 /* Gain Table Write Data 3 */
#define REG_GAIN_TABLE_READ_DATA1		 0x134 /* Gain Table Read Data 1 */
#define REG_GAIN_TABLE_READ_DATA2		 0x135 /* Gain Table Read Data 2 */
#define REG_GAIN_TABLE_READ_DATA3		 0x136 /* Gain Table Read Data 3 */
#define REG_GAIN_TABLE_CONFIG			 0x137 /* Gain Table Config */

#define REG_GM_SUB_TABLE_ADDRESS	     0x138 /* Gm Sub Table Address */
#define REG_GM_SUB_TABLE_GAIN_WRITE		 0x139 /* Gm Sub Table Gain Word Write */
#define REG_GM_SUB_TABLE_BIAS_WRITE		 0x13A /* Gm Sub Table Bias Word Write */
#define REG_GM_SUB_TABLE_CTRL_WRITE		 0x13B /* Gm Sub Table Control Word Write */
#define REG_GM_SUB_TABLE_GAIN_READ		 0x13C /* Gm Sub Table Gain Word Read */
#define REG_GM_SUB_TABLE_BIAS_READ		 0x13D /* Gm Sub Table Bias Word Read */
#define REG_GM_SUB_TABLE_CTRL_READ		 0x13E /* Gm Sub Table Control Word Read */
#define REG_GM_SUB_TABLE_CONFIG			 0x13F /* Gm Sub Table Config */

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25

开FIR可能会影响时序

RX接收数据解析

hdl-2016-r2版本


hdl-2018-r1版本

hdl-2019-r2版本


要注意，

• 2016-r2和2018-r1的lvds_if里，解数据的方式不一样,IDDR的使用方式也不一样
• 2018-r1和2019-r2的lvds_if里，解数据的方式不一样

其他

FDD
Frequency division duplex in which transmit and receive signals can be present at the same time but use different frequencies.
是不是意味着同时进行收发的时候，收发频率不能相同？

AD9361上电后进入休眠模式（为了降低功耗），在AD9361能正常运行之前，它的时钟必须已经使能且初始化校正完成。

ADI官方论坛网友提供的AD9361初始化